Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    marriage and love

    np_max-act · gemini-2.0-flash

    The neuron activates on words expressing promises or commitments (e.g. “promised,” “give,” “support,” “to”)—i.e. the language of vows.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     tận
    -0.07
     Elli
    -0.07
     venom
    -0.07
     Sons
    -0.06
    eil
    -0.06
    ecimal
    -0.06
     mosaic
    -0.06
    尿
    -0.06
    	root
    -0.06
    ेटर
    -0.06
    POSITIVE LOGITS
     Jaguars
    0.06
     Jub
    0.06
     MY
    0.06
    .HttpStatus
    0.06
     classrooms
    0.06
     knit
    0.06
    \Extension
    0.06
    ^{-
    0.06
     Budget
    0.06
     airplanes
    0.06
    Act Density 0.039%

    No Known Activations